Generating a multimedia document, such as a yearbook or slideshow, that includes a set of images that is representative of different demographic groups (e.g., a group of individuals in their 20s, a group of individuals in their 30s, etc.) within a large group of members (e.g., employees at a company) can be a difficult task. For example, it takes a lot of effort and time in creating an end of the year slideshow for an organization (e.g., company, non-profit organization) that includes pictures of members of the organization (e.g., employees of company). The creator(s) of the slideshow may want to ensure that each demographic group (e.g., groups based on age, race, gender) are not overly or inadequately represented in comparison to the other demographic groups in order to provide fairness and increase member enjoyment. For example, the creator(s) of the yearbook may want to ensure that the slideshow does not include many images of a particular demographic group (e.g., twenty pictures of males) while only including a single image of another class demographic group (e.g., a single picture of a female). However, the creator(s) of the slideshow may have hundreds or thousands of images of hundreds of members of the organization to choose to include in the slideshow. As a result, it may be overwhelming for the creator(s) of the slideshow to ensure that each member of the organization is adequately represented in the slideshow and that there are not some members that are overly represented in the yearbook.
Currently, a multimedia document, such as a slideshow, may be generated by selecting an image for inclusion in the slideshow, where the image has associated facial detection information. A face location is determined in the selected image based on the facial detection information and the selected image is cropped based on the determined face location to generate a cropped image depicting the included face. The cropped image is inserted into a slide associated with the slideshow. However, there is not currently an approach in ensuring a specific distribution of images in the slideshow for various demographic groups, such as ensuring a particular demographic group is not overly or insufficiently represented in the slideshow.
Furthermore, a multimedia document, such as a yearbook, may be designed and created utilizing a facial recognition module for identifying a subject in a photo and tagging the photo with an identity of the subject. Identifying the subject may include comparing the photo to a plurality of previously tagged photos to identify the subject. However, there is not currently an approach in ensuring a specific distribution of images for demographic groups in the yearbook. For example, there is not currently an approach in ensuring that each demographic group has a minimum number of images in the yearbook. Neither is there currently an approach in ensuring that there is a certain percentage of images out of the total number of images in the yearbook for a particular demographic group.
Therefore, there is a need in the art for ensuring a desired distribution of images for various demographic groups in a multimedia document (e.g., yearbook, slideshow, video, website).